LCSH Tools
LCSH Tools is an ecosystem of AI-powered applications that help librarians and catalogers assign Library of Congress Subject Headings (LCSH) to bibliographic works. It combines AI with real-time validation against the Library of Congress authorities database to generate standards-compliant MARC records.
Three Tools, One Ecosystem
Cataloging Assistant (Flagship)
A Progressive Web App with a guided 3-step wizard. Enter bibliographic info — title, author, abstract, table of contents, or upload book covers — get AI-powered LCSH suggestions validated against the Library of Congress, and export MARC records. Supports multiple AI providers (OpenAI, Google Gemini, DeepSeek, Groq, and more) and works offline.
Chrome Extension
Get LCSH recommendations directly in your browser sidebar. Powered by Google Gemini, it validates suggestions against id.loc.gov in the background and generates MARC records — all without leaving your tab.
MCP Server
Give Claude and other AI assistants direct access to the Library of Congress authorities database. Search LCSH and LCNAF through the Model Context Protocol with three specialized tools. Available on PyPI as cataloger-mcp-server.
Key Features
- Validated Accuracy — Every AI suggestion is checked in real time against the Library of Congress authorities database with similarity scores.
- Standards-Compliant — Generates properly formatted MARC records (650, 600, 610) with correct indicators and subfields, ready for ILS import.
- Privacy-First — Zero-knowledge architecture. All data stays in your browser. API keys are stored locally and never transmitted to any server except your chosen AI provider.
- Multi-Provider AI — Bring your own API key from OpenAI, Google Gemini, DeepSeek, Groq, Perplexity, OpenRouter, or any OpenAI-compatible endpoint.
Research
The methodology is described in: Tang, K. L. & Jiang, Y. (2025). “Better Recommendations: Validating AI-generated Subject Terms Through LOC Linked Data Service.” Available at arXiv:2508.00867.